Abhinay,
Thanks for that. I did look thought the thesis and there's a lot of
useful information. Unfortunately though I couldn't find anything to
address the specific issue (unless I missed it - I'm starting to go
cross-eyed now).
I've found much online comparing different registration/normalization
methods but nothing that compares functional->functional template and
coregistered functional-structural->structural template.
The only references I've found mention that structural normalization
will be more accurate than functional as it contains more information.
But its not clear, for example, whether coregistration of PET-MR and
then a transform (derived from MR->MR template) will introduce more
errors than going straight from PET->PET template.
Maybe its a bit too specific.
Joel
Abhinay Joshi wrote:
> Type " John Ashburner" in google, and try the second hit or look for
> "John Ashburner's thesis", read through the chapters, you will find
> the answer.
>
> Abhinay
>
>
> On 2/28/07, *Joel Dunn* <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> Hi,
> It seems 'common sense' to assume that spatially normalizing PET
> images via
> the normalization of the subject's coregistered MRI to the MRI
> template is
> more accurate than direct normalization of the PET to the PET
> Template.
>
> But does anyone know of any references that support this?
>
> (I've spent 4 hours on PubMed & Web of Knowledge without much
> luck...)
> Joel
>
>
>
>
> --
> Abhinay D. Joshi
> UT Southwestern
> E06.122 MC 9061
> Dallas, TX 75390
> P -214-648-9265
>
> Principal Imaging Scientist
> Avid Radiopharmaceuticals, Inc.
> 3624 Market Street, 5th Floor
> Philadelphia, PA 19104
--
-----------------------------------------------
Joel Dunn
Research Assistant
PET Imaging Centre
Guy's, King's and St Thomas' School of Medicine
Lower Ground Floor, Lambeth Wing,
St Thomas' Hospital, Lambeth Palace Road,
London SE1 7EH UK Tel: 020 7188 7445
Email: [log in to unmask] Fax: 020 7620 0790
-----------------------------------------------
|